Partner – Microsoft – NPI (cat= Spring)
announcement - icon

Azure Spring Apps is a fully managed service from Microsoft (built in collaboration with VMware), focused on building and deploying Spring Boot applications on Azure Cloud without worrying about Kubernetes.

And, the Enterprise plan comes with some interesting features, such as commercial Spring runtime support, a 99.95% SLA and some deep discounts (up to 47%) when you are ready for production.

>> Learn more and deploy your first Spring Boot app to Azure.

You can also ask questions and leave feedback on the Azure Spring Apps GitHub page.

1. Overview

The Spring IoC container creates and manages Spring beans, which serve as the core of our application. Creating an instance of a bean is identical to creating objects from plain Java classes. However, generating several beans of the same class can be challenging.

In this tutorial, we’ll learn how to use annotations in the Spring framework to create multiple beans of the same class.

2. Using Java Configuration

This is the simplest and easiest way to create multiple beans of the same class using annotations. In this approach, we’ll use a Java-based configuration class to configure multiple beans of the same class.

Let’s consider a simple example. We have a Person class that has two class members, firstName and lastName:

public class Person {
    private String firstName;
    private String lastName;

    public Person(String firstName, String secondName) {
        super();
        this.firstName = firstName;
        this.lastName = secondName;
    }

    @Override
    public String toString() {
        return "Person [firstName=" + firstName + ", secondName=" + lastName + "]";
    }
}

Next, we’ll construct a configuration class called PersonConfig, and define multiple beans of the Person class inside it:

@Configuration
public class PersonConfig {
    @Bean
    public Person personOne() {
        return new Person("Harold", "Finch");
    }

    @Bean
    public Person personTwo() {
        return new Person("John", "Reese");
    }
}

Here, @Bean instantiates two beans with ids the same as the method names, and registers them within the BeanFactory (Spring container) interface. Next, we can initialize the Spring container, and request any of the beans from the Spring container. This strategy also makes it simple to achieve dependency injection. We can directly inject one bean, say personOne, into another bean of the same type, say personTwo, using autowiring.

The limitation of this approach is that we need to manually instantiate beans using the new keyword in a typical Java-based configuration style.  Therefore, if the number of beans of the same class increases, we need to register them first, and create beans in the configuration class. This makes it a more Java-specific approach, rather than a Spring-specific approach.

3. Using @Component Annotation

In this approach, we’ll use the @Component annotation to create multiple beans that inherit their properties from the Person class. First, we’ll create multiple subclasses, namely PersonOne and PersonTwo, that extend the Person superclass:

@Component
public class PersonOne extends Person {

    public PersonOne() {
        super("Harold", "Finch");
    }
}
@Component
public class PersonTwo extends Person {

    public PersonTwo() {
        super("John", "Reese");
    }
}

Next, in the PersonConfig file, we’ll use the @ComponentScan annotation to enable component scanning throughout the entire package. This enables the Spring container to automatically create beans of any class annotated with @Component:

@Configuration
@ComponentScan("com.baeldung.multibeaninstantiation.solution2")
public class PersonConfig {

}

Now we can just use the PersonOne or PersonTwo beans from the Spring container. Everywhere else, we can use the Person class bean. The problem with this approach is that it doesn’t create multiple instances of the same class. Instead, it creates beans of classes that inherit properties from a superclass.

Therefore, we can use this solution only in situations where the inherited class doesn’t have any additional properties defined. Moreover, the usage of inheritance increases the overall complexity of the code.

4. Using BeanFactoryPostProcessor

The third and final approach utilizes a custom implementation of the BeanFactoryPostProcessor interface for creating multiple bean instances of the same class.

This can be achieved using the following steps:

  • Creating a custom bean class and configuring it using the FactoryBean interface
  • Instantiating multiple beans of the same type using the BeanFactoryPostProcessor interface

4.1. Custom Bean Implementation

To understand this approach better, we’ll extend the same example further. Suppose there’s a Human class that’s dependent upon multiple instances of the Person class:

public class Human implements InitializingBean {

    private Person personOne;

    private Person personTwo;

    @Override
    public void afterPropertiesSet() throws Exception {
        Assert.notNull(personOne, "Harold is alive!");
        Assert.notNull(personTwo, "John is alive!");
    }

    /* Setter injection */
    @Autowired
    public void setPersonOne(Person personOne) {
        this.personOne = personOne;
        this.personOne.setFirstName("Harold");
        this.personOne.setSecondName("Finch");
    }

    @Autowired
    public void setPersonTwo(Person personTwo) {
        this.personTwo = personTwo;
        this.personTwo.setFirstName("John");
        this.personTwo.setSecondName("Reese");
    }
}

The InitializingBean interface invokes the afterPropertiesSet() method to check whether BeanFactory has set all the bean properties, and satisfied other dependencies. Additionally, we’re injecting and initializing two Person class beans, personOne and personTwo, using setter injection. Next, we’ll create a Person class that implements the FactoryBean interface.

A FactoryBean acts as a factory for creating other beans within the IoC container. This interface is intended to create more instances of the bean that implements it. In our case, it generates instances of the type Person class, and configures it automatically:

@Qualifier(value = "personOne, personTwo")
public class Person implements FactoryBean<Object> {
    private String firstName;
    private String secondName;

    public Person() {
        // initialization code (optional)
    }

    @Override
    public Class<Person> getObjectType() {
        return Person.class;
    }

    @Override
    public Object getObject() throws Exception {
        return new Person();
    }

    public boolean isSingleton() {
        return true;
    }

    // code for getters & setters
}

The second important thing to notice here is the use of the @Qualifier annotation, which contains the names or bean ids of multiple Person types at the class level. There’s a reason behind using @Qualifier at the class level, in this case, which we’re going to see next.

4.2. Custom BeanFactory Implementation

Now we’ll use a custom implementation of the BeanFactoryPostProcessor interface.

Any class that implements BeanFactoryPostProcessor is executed before any Spring bean gets created. This allows us to configure and manipulate the bean lifecycle.

The BeanFactoryPostProcessor scans all the classes annotated with @Qualifier. Furthermore, it extracts names (bean ids) from that annotation, and manually creates instances of that class type with the specified names:

public class PersonFactoryPostProcessor implements BeanFactoryPostProcessor {

    @Override
    public void postProcessBeanFactory(ConfigurableListableBeanFactory beanFactory) throws BeansException {
        Map<String, Object> map = beanFactory.getBeansWithAnnotation(Qualifier.class);
        for (Map.Entry<String, Object> entry : map.entrySet()) {
            createInstances(beanFactory, entry.getKey(), entry.getValue());
        }
    }

    private void createInstances(ConfigurableListableBeanFactory beanFactory, String beanName, Object bean) {
        Qualifier qualifier = bean.getClass().getAnnotation(Qualifier.class);
        for (String name : extractNames(qualifier)) {
            Object newBean = beanFactory.getBean(beanName);
            beanFactory.registerSingleton(name.trim(), newBean);
        }
    }

    private String[] extractNames(Qualifier qualifier) {
        return qualifier.value().split(",");
    }
}

Here, the custom BeanFactoryPostProcessor implementation gets invoked once the Spring container is initialized.

Next, to keep things simple, we’ll use a Java configuration class to initialize the custom, as well as BeanFactory, implementations:

@Configuration
public class PersonConfig {
    @Bean
    public PersonFactoryPostProcessor PersonFactoryPostProcessor() {
        return new PersonFactoryPostProcessor();
    }

    @Bean
    public Person person() {
        return new Person();
    }

    @Bean
    public Human human() {
        return new Human();
    }
}

The limitation of this approach lies in its complexity. Moreover, it’s not encouraged to use, since it’s not the natural way of configuring beans in a typical Spring application. Despite the limitations, this approach is more Spring specific, and serves the purpose of instantiating multiple beans of similar type using annotations.

5. Conclusion

In this article, we learned about instantiating multiple beans of the same class using Spring annotations through three different approaches. The first two approaches were simple, Java-specific ways to instantiate multiple Spring beans. The third one was a bit tricky and complex, but it serves the purpose of bean creation using annotations.

As always, the source code for the examples is available on GitHub.

Course – LS (cat=Spring)

Get started with Spring and Spring Boot, through the Learn Spring course:

>> THE COURSE
res – REST with Spring (eBook) (everywhere)
Comments are open for 30 days after publishing a post. For any issues past this date, use the Contact form on the site.